# A Closer Look at Dynamic Scene Graph Generation In the Era of Multimodal Large Language Models

arXiv: https://arxiv.org/abs/2503.15846 · HTML: https://mlsanity.com/paper/2503.15846/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 28 of its 47 references that are in the map.
- First posted on arXiv 20 Mar 2025.

## Placed in

- Multimodal LLMs / vision-language models / [VLM benchmarks and visual reasoning](https://mlsanity.com/mllm/vlm-benchmarks-and-visual-reasoning/)

## Builds on (most cited first, 28 of 28)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis](https://mlsanity.com/paper/2405.21075/) (2024)
- [Scene Graph Generation by Iterative Message Passing](https://mlsanity.com/paper/1701.02426/) (2017)
- [Neural Motifs: Scene Graph Parsing with Global Context](https://mlsanity.com/paper/1711.06640/) (2017)
- [Is ChatGPT Good at Search? Investigating Large Language Models as Re-Ranking Agents](https://mlsanity.com/paper/2304.09542/) (2023)
- [Unbiased Scene Graph Generation from Biased Training](https://mlsanity.com/paper/2002.11949/) (2020)
- [Visual Relationship Detection with Language Priors](https://mlsanity.com/paper/1608.00187/) (2016)
- [Graph R-CNN for Scene Graph Generation](https://mlsanity.com/paper/1808.00191/) (2018)
- [Action Genome: Actions as Composition of Spatio-temporal Scene Graphs](https://mlsanity.com/paper/1912.06992/) (2019)
- [GPS-Net: Graph Property Sensing Network for Scene Graph Generation](https://mlsanity.com/paper/2003.12962/) (2020)
- [AGQA: A Benchmark for Compositional Spatio-Temporal Reasoning](https://mlsanity.com/paper/2103.16002/) (2021)
- [Spatial-Temporal Transformer for Dynamic Scene Graph Generation](https://mlsanity.com/paper/2107.12309/) (2021)
- [Target Adaptive Context Aggregation for Video Scene Graph Generation](https://mlsanity.com/paper/2108.08121/) (2021)
- [Unbiased Scene Graph Generation in Videos](https://mlsanity.com/paper/2304.00733/) (2023)
- [Symbolic Replay: Scene Graph as Prompt for Continual Learning on VQA Task](https://mlsanity.com/paper/2208.12037/) (2022)
- [Exploiting Long-Term Dependencies for Generating Dynamic Scene Graphs](https://mlsanity.com/paper/2112.09828/) (2021)
- [(2.5+1)D Spatio-Temporal Scene Graphs for Video Question Answering](https://mlsanity.com/paper/2202.09277/) (2022)
- [SOK-Bench: A Situated Video Reasoning Benchmark with Aligned Open-World Knowledge](https://mlsanity.com/paper/2405.09713/) (2024)
- [Unbiased Scene Graph Generation via Two-stage Causal Modeling](https://mlsanity.com/paper/2307.05276/) (2023)
- [Cross-Modality Time-Variant Relation Learning for Generating Dynamic Scene Graphs](https://mlsanity.com/paper/2305.08522/) (2023)
- [Motion-aware Contrastive Learning for Temporal Panoptic Scene Graph Generation](https://mlsanity.com/paper/2412.07160/) (2024)
- [Conformal Prediction and MLLM aided Uncertainty Quantification in Scene Graph Generation](https://mlsanity.com/paper/2503.13947/) (2025)
- [FloCoDe: Unbiased Dynamic Scene Graph Generation with Temporal Consistency and Correlation Debiasing](https://mlsanity.com/paper/2310.16073/) (2023)
- [Unbiased Video Scene Graph Generation via Visual and Semantic Dual Debiasing](https://mlsanity.com/paper/2503.00548/) (2025)

## Abstract (arXiv)

Dynamic Scene Graph Generation (DSGG) aims to capture objects and their evolving relations in videos. Despite recent progress, the practicality and quality of generated scene graphs remain limited compared to the rapid advances in Multimodal Large Language Models (MLLMs). In this work, we revisit DSGG from two fundamental perspectives: task setup and model design. From the task setup perspective, we identify two key limitations of the current recall-oriented evaluation protocol: (i) a severe precision-recall trade-off, and (ii) uninformative and redundant relation generation. To better assess practical usefulness, we introduce five additional metrics that provide a more comprehensive evaluation of the quality of generated dynamic scene graphs. From the model design perspective, we explore directly using MLLMs for scene graph generation and establish a strong MLLM-based DSGG baseline through three design changes. First, we replace the conventional bottom-up pipeline with a top-down reason-then-locate strategy. Second, we reformulate frame-wise dynamic graphs as Temporal Relation Set (TRS) prediction, improving both efficiency and performance. Third, we introduce Importance-Aware Finetuning (IAF) to encourage more relevant and diverse relation generation. Extensive experiments on Action Genome, VidVRD, and PVSG show that our approach consistently achieves state-of-the-art performance.